Friday Facts #275 - 0.17 Science changes

[V453000]

First of all, good work! I really like these changes! And that means actually a lot, since i never liked the science before ever lol.

Ofc I still have some complaints anyways
1.) As thue already said I also was surprised the blue science crafting time wasn't adjusted to tier 2 science - i’d suggest like 16s? it should be the same for military and blue science anyways, and black science really misses some additional assemblers in its chain, so a slower crafting time seems fine for the final craft.
2.) Purple science is now very steel heavy, using 25 steel per 3. However i still like the recipe. Map gen may require every more iron boosted balance now - i don't know any solution to it tho, seems fine
3.) Purple science is MUCH easier than Yellow from a complexity view - Taking away the red engines which are the most complex part due to the addition spaghetti with the lube, now the difficult recipes of blue chip and bots are both is utility science. - I think this is fine tho, but it definitely fails to offer alternative routing imo.
4.) Removing Black science from the silo is an odd choice imo. it makes black science pretty much useless now and required in very few numbers only. Taking into account the science also got less costly, most people will barely ever need more than 1 assembler of each in its production chain. If you want to end the game, you should master the production of ALL the sciences IMO.
5.) I liked the Idea of the escape pod a lot and I found that a good solution to end the game. With it being abandoned, I cannot really understand the removal of the satellite from its prerequisites. If you are worried about people launching a rocket without a satellite, you can still show the cargo slot while building the rocket and warn the player with a massage when he tries to launch the rocket without any cargo first, if he really plans to do that and it won't win him the game. It reduced complexity where it is not needed IMO

Thank you, I appreciate you saying that!

1) I do agree that it should be lower. Maybe Military crafting time could be higher instead as you don't need to have it run nonstop.
2) Steel heavy is not a problem, in fact using the more sophisticated ingredient as steel compared to iron plate in a high-tier science pack is good.
3) The two sciences (purple and yellow) are not (meant to be) a completely symmetrical balance. Each of them is different and for different uses. Some parts of the recipe chain are more complicated and some less, but in total I would not say that for example specifically electric engines are that complex to make. Together with the flying frames it is a bunch, but the production science pack has its own set of difficulties when you want to set up the production for it. The "total time required for crafting" is a very interesting number here which shows that the total number of machines should be roughly the same for both science packs, and yellow also needs more oil. The main thing for me is that whether you go for purple or yellow first is completely up to you, and whether you treat one as more useful over the other is entirely based on the situation, the map, the settings, your house rules, and your playstyle. If you're someone who uses roboports ASAP, you probably aren't so thrilled about mk2 power armor or fusion reactor with mk2 personal roboports, and vice versa. If you play without biters you're more likely to use the purple pack first but again could be counteracted if you also love the mk2 armor due to exoskeletons. The options are there and if you choose one path it feels like it's self-justified and you're not screwing yourself by your choice.
4) It's expectable that if you play with biters, they will put enough pressure on you that you will gladly invest into military. The split from it being mandatory also means that it's further underlined that it is something optional and military-only. IF the rocket silo pre-requisite would be something substantial like Power armor mk2, then I'd say that's a nice pre-requisite useful for everybody and the military science pack is just-another-recipe you use even in non-military games. Which is basically what will happen anyway with anybody except speedruns because you want the mk2 armor and fusion reactor regardless.
5) I do believe just launching a rocket is enough, you can already find excuses like "it's a spaceship" and that the player could enter it regardless. It also leaves it open for the player to figure out if they left the planet or not which find really good. The whole "you crashed on an alien planet" obviously brings the cliche idea of "you are obviously trying to escape it", but eventually you get so good at automation (and genocide) that you decide that this is your planet now. For example - the interpretation there is for you as a player to fill in, and I think that's nice. Don't need to shove it down your throat.

[Lizzy]

You should allow a player to be the rocket payload.

...then, if you click "Continue" in the victory screen, you drop-off in a far-off spot in the map with your inventory only. Preferably without a way of returning.
https://tvtropes.org/pmwiki/pmwiki.php/Main/NewGamePlus

The only thing I would disagree with here are the productivity modules for production packs: All the other things are going to be used sooner or later, but it is very well possible to never use prod modules, so this feels like you are trying to force us to use them(especially Level 1, which I never saw used anywhere).

The thing is, using a thing in a science pack isn't really forcing you to do anything, just hinting that it might be a good idea. If you play without productivity modules then you are deliberately restricting yourself.

I don't use low-level modules, even though I know they make sense to optimize my factory, because they're a pain to swap over. Even high-level ones are a nightmare to deploy partially. If modules are supposed to be a larger part of the game, they need a lot of QoL improvements for adding/removing/replacing them for large parts of the factory. Destroying and rebuilding assembly machines, the current easiest method, generates LOTS of spillover into the logistics network.

Now on top of this, i think satelites should make radars semi obsolete after a few of them in space.
What i think, the satelites should not reveal new black area, but ocationally remove the fog of war of the whole map.
The more satelites in orbit, the longer the fog is gone, untill eventually it will be completely gone.
I believe this will be a really good reward for keep sending rockets out there with satelites.

Hey, I love this! It would be a way bigger incentive for me than infinite research.

Uranium:

I'd like to suggest that uranium also be made relevant. As it stands, it's a cool resource that frequently serves little purpose - especially if you're megabasing with solar + biters off.

I agree with Neuromaster that I do wish uranium had more uses. It's a cool part of the game, and I feel like (because it hasn't had as long to develop) it isn't really at its peak potential. I would say a fuel cell would be a rather high cost for some of the science packs, but I wonder if there isn't room for a little uranium somewhere in the rocket construction process.

I'd like to add my voice to this chorus. Uranium ore is fun, I'd love to see it prancing along more sections of the factory in the late game.

Ooooh, one more thing... Can we get even number of science packs somehow... Because reasons. VERY important reasons.

You could just fill out that last half-belt with some amulet-like item. Say, like raw Uranium. Or modules. Or fishes. Yeah, definitely fishes.

Also, I now have a new hobby: Laughing at people calling beacons "magic" even as they somehow hand-craft electric circuits that they then use to hand-craft fully functional six-axis robots. Let's not even get into the nonsense that is assembling machines, which are a 3x3 that can have various resources pumped in to any of 8 potential different locations and still not only work perfectly fine but requires nothing more than a sternly worded "Now you're going to produce this belt for the next 10000 years and I want no complaints from you!" in order to magically have it function perfectly for the rest of time.

Yeah, I thought the same. Conveyor belts that require no upkeep or energy, and don't crumple and fall over when overfull? Completely unrealistic player and chest inventory sizes, like any self-respecting modern game? [sarcasm]No magic in this here game, nope![/sarcasm]

Apart from the ridiculously fast hand-crafting speed (or the slow assembling machines), it is in theory possible to hand-craft a fully functional robot. It takes time and effort, but it is entirely possible.

You forgot walking like crafting, completely freeing up available (imaginary) work-desk space while working on a large hand-crafting batch, and so on. No game is realistic. No game is 100% reality-compliant - not even educational science simulators.

I love Factorio because (1) it is fun, and (2) it feels sufficiently engineering-like.

Some people need to ignore parts of the game to accomplish criterion (2). I'm not one of these people, but whatever floats your boat. I personally interpret the "unrealistic" parts of the game as a separate universe with separate laws. And even though I defend that, I have a really hard time accepting Factorio science packs and Factorio research labs entirely. "How on Earth* is research something you can automate in a completely unattended way? Ugh, let me just plop this science blueprint and move on."

*off Earth, I guess.

Anyway, moving from a Watsonian PoV to a Doylist PoV: just in the previous FFF there was a lengthy explanation on how going for more realism would result in a worse game. (In that particular case, too CPU-intensive to stand larger bases.)

[Mike5000]

I'd love to reply to all of you, hopefully I will find time and energy for it later, but just to clarify one very wrong thing:

The personal roboport in yellow tier you are refering to is Personal roboport mk2. Personal roboport mk1 is on blue science.

Perhaps you could correct the FFF where it says "Utility science pack alone now unlocks ... Personal roboport, Power armor Mk2, Fusion reactor ...".

[abregado]

Why have recipes produce 2x and 3x science packs? Why not make 1x pack but the research cost 2x or 3x cheaper? Seems more straightforward than recipes producing multiples

It just seems an unnecessary extra multiplier factor to use in your head when computing research cost from resources that can be avoided. I have nothing against other multi output recipes like bullets.

Doing this also exacerbates the "shunting problem" which requires thinking about logistical problems in new ways. This is not why we changed it, just a nice side effect.

I feel that this "hidden meta complexity" adds a lot of depth to the game for new and veteran players, and smoothing it out like this will help the onboarding a lot without "making the game for babies".

[abregado]

If you are going to call the first two science packs "Automation" and "Logistics", consider how this affects the new player experience together with the Automation and Logistics research. Automation 1/2 only requires Automation science packs, which makes sense, but then there is Logistics 1 which doesn't require Logistics science packs, while Logistics 2 does. This could cause some confusion for players who see the tech tree and science packs for the first time.

As your resident dev Representative of the new player, consider your concern noted. The tech tree has seen some changes in general that should help with this, but we are also doing new player focus testing and it will be something Ill keep an eye out for.

[abregado]

Anyway, moving from a Watsonian PoV to a Doylist PoV: just in the previous FFF there was a lengthy explanation on how going for more realism would result in a worse game. (In that particular case, too CPU-intensive to stand larger bases.)

Thanks for the link, I now have two new ways to use the word "diegetic".
Many successful game designers can be quoted saying something akin to: "Dont let realism get in the way of a good design, but never let a good design get in the way of consistency".

[Dry Hairy Tree]

I REALLY like most of the suggested changes but agree with the following:

Beacons should be out. Relegate the silly things to mods.

Rail isn't the right ingredient.

Uranium in high tech makes sense.

It was a lot to take in. You've done a lot of thinking it over. Grateful as always.

[Reika]

I quite like all the science pack changes.

And count my vote in favor of adding the pipe to Science Pack 2/Logistic Science/Green Science.

[Oktokolo]

Me too likes the new facts.
I would like having construction bots before petrochem and would drop the packs for directly feeding much more ingredients to specialized labs. But early bots work just fine for me and the science tree i will probably mod myself anyway - if i ever get enough spare time to do so while still playing the game...
Good idea to find more uses for stone and lowdensity structs.

Beacons should be out. Relegate the silly things to mods.

If axing them makes the engine less complex, they should definitely consider it.
Otherwise there is no reason to remove them. They are so lategame, that the risk of player confusion is negligible.

[cd579]

Not sure if anyone has said this before, but as a longtime factorio player, I still struggle with the damage values against biters.

"Also, it means for example Tank cannon shell upgrades do not have 7 tiers but just 2 (Physical projectile damage 5+ and Weapon shooting speed 5+), but instead they are expensive and very impactful."

I still have no clue what 5+ damage or 5+ weapon shooting speed means. I think newbies and even long time players might benefit from a redo on this part, especially considering your verbiage overhaul for the rest of the game. Possibly a "50% increase" instead as we know directly what that means. I know this gets difficult to reconcile as the biter difficulty increases, but I have confidence in your team. I am VERY excited for all the changes you have planned for the new release!

Thank you for your intense effort to make the best game!

[morsk]

Excited for the new recipes. I tend to quit bases around the transition from blue to purple science, as the scale gets huge, but I don't have exoskeletons and everything feels so tedious. With purple vs. yellow more of a choice, I can finally get yellow first, and then do purple after I can actually move. I like the robot frames as part of yellow science. That's a very complex recipe, especially for how early in the game it's learned. It will be satisfying to design an assembly line good enough for initial bot production, and for keeping around to reuse for yellow science hours later. Especially if it's integrated with the same engine production blue science uses.

Pipes in green science feels unnecessarily annoying. Green science already has the inserter to teach players a recipe with 3 ingredients. And it's an interesting one, as it uses gears and circuits itself. Adding a pipe to the end is just hassle with more belts. I don't think it will help building oil either. That takes a mini-mall that stocks a few hundred pipes, along with underground pipes. At some point players need to learn to build little malls, and hopefully the campaign itself can teach this.

[gacekssj4]

Just my opinion, but you can do both about since packs and put them into Tiers to indicate their complexity or order

For example:

factorio_tier_sience.jpg (45.15 KiB) Viewed 5729 times

[WIZ4]

This postcard lacks continuation.

atomic-bomb.gif (6.95 MiB) Viewed 5718 times

[FasterJump]

What I like:
-Reduced cost per science pack 3
-Merging military research
-New name of science packs. Except maybe yellow, why not name it simply "Tech"? I'm fine with "Utility" anyway.
-Using solid fuel as ingredient for sp3. It's a smart move to make use of the excess oil.
-Use of prod I modules in production science pack.
-I like that you consider prod III modules to calculate the consumption of a base, and tweak the ore map generation accordingly

What I dislike:
-Use of low density structures and robot frames in utility science pack.
-Winning the game without satellite: I don't mind, but a pop-up warning about the empty rocket would have make more sense imo.

What I hate:
Rails in sp3 recipe. That makes no sense. A rail is 2x2 tiles. I can't even figure how rails are internally used by the assembler. 30 of them? No way.
On another side, wires in purple science makes a bit more sense. I can figure that the wires are used by the assembler to connect stuff and end up with "science".
I'm not fan of the high throughput challenge about those 30 wires, because it forces you to use direct insertion. But since it adds more diversity to the game, and fits in the "Hi-tech" theme of the yellow science, I suppose it's fine.
Having more of this mechanic in the game doesn't seems like a good idea to me.
Anyway it's just my 2 cents, maybe it would be less bad that I imagine.

Bunch of quotes about sp3 craft time

24 seconds for "chemical science pack"s are too much. Considering you get 3 out of "Production science pack"s and "Utility science pack"s and both only take 21 seconds.

what about the time it takes to produce blue packs - this is non linear

The crafting time for blue science is too long. I know this hasn't changed; I think it should because it was always weird. All other science packs (except space) take 5, 6 or 7 seconds to craft after taking batch size into account. Blue is freakin' 12, double the average. Why is that necessary?

Chemical science packs
As others commented the 24s crafttime feels really weird compared to the other science craft times.

1.) As thue already said I also was surprised the blue science crafting time wasn't adjusted to tier 2 science - i’d suggest like 16s?

1) I do agree that it should be lower.

Oh please YES yes and yes!

[featherwinglove]

Beacons are "magic" because they somehow give you stuff without you having to pay for those materials. With iron plates and so on, that makes sense (perhaps you only need 90% of a plate, so the rest is waste, but in this case will be used for later stuff), but with circuits for example? How do you use 90% of a fully assembled circuit board? Even better, how do you assemble 9 times the remaining 10% to add up to another 90%? (numbers pulled out of some biter's remains)

You seem to be getting beacons mixed up with the productivity effect.

The goal isn't necessarily to tie the pack name to the things that make up the pack. (Do low density structures provide some kind of "utility"? How are red circuits connected to "chemistry"?) It's more related to game objectives of that time. The post gave some good explanations for why they wanted certain ingredients in the science packs and not others - I think taking issue with those explanations is fine, but the thematic element is secondary to the goal of "we want players to be considering these objects/using these ideas by this point in the game".

As some people have pointed out, it's not like the item in the recipe - rails - has to be a thing which actually does production.
...
As such I think it makes perfect sense to include rails in the production science pack, because starting to use rails really does result in much higher production numbers, in a roundabout way.

It isn't just the way rails don't relate to "production", they don't make sense as a consumable in a science pack because of how they're a passive, low-wear infrastructure item. If you want something train-related, I think something in the vehicle or signal hardware would make much more sense than simple rails.

But I do not like the Idea of having all 6 ingredients slots free in assembling machine.

In the FFF that brought it up, pretty much the entire community spoke out against it and Wube seemed to simply ignore the outcry. It's definitely a strange outlier for a team that's at least been transparent and interacted with the community on everything.

Are you referring to the catastrophe following FFF#266? I don't like anything in that one, but IMHO, the removal of AM ingredient limits is by far the least offensive of the announced changes. Wube's behaviour following it actually has me warning my friends off this game (that included not updating the roadmap thread for three months and then deleting my post when I did. And that's not the only part of it that is Blizzcon 2018 level bullshit.)

[AngledLuffa]

It's going to take so much work to upgrade my megabase to the 0.17 science packs. I can't wait!

Please do make it feasible to upgrade existing bases, of course

[SirLANsalot]

Great FFF, but just one thing.

Military Science. Grenades are rarely used....like if you wanted to change the recipe to reflect USEFUL things, grenades are not one of them. They are far too clumsy to be useful in combat at all, and while good for clearing trees its very wasteful. Burning trees is more fun and it does it on its own, just a little fire and, no more forest.

As for everyone else bitching about Rails being added to the packs. You ALWAYS want to automate rail production, because you are going to need a TON of them mid to late game as you push outwards. Resources do not last forever (unless you use angles infinite ores, but that still won't be at full production) meaning you need to push further and futher away from your base. You will need a LOT of rails and you will be thanking your early game self, for automating that production line.
Now its a requirement, which also means Stone now becomes ACTUALLY USEFUL. Before, stone was just something you needed once in a while for walls but mostly just rails. Draining one or two stone locations provided more then enough stone for...well the rest of the game. Now you will need to secure stone spots as well as Iron and Copper spots as the game moves on. Coal usually gets replaced soon and the grenade requirement on military doesn't really take that much coal, so usually its just needed for Plastic. Having a few Coal spots with a few small trains is more then enough to feed that requirement.

Very excited and looking forward to playing .17 once released to beta.

[featherwinglove]

Resources do not last forever (unless you use angles infinite ores...

There are so many infinite ore mods, saying this is just

Great FFF, but just one thing.

Military Science. Grenades are rarely used....like if you wanted to change the recipe to reflect USEFUL things, grenades are not one of them. They are far too clumsy to be useful in combat at all, and while good for clearing trees its very wasteful. Burning trees is more fun and it does it on its own, just a little fire and, no more forest.

As for everyone else bitching about Rails being added to the packs. You ALWAYS want to automate rail production, because you are going to need a TON of them mid to late game as you push outwards. Resources do not last forever (unless you use angles infinite ores, but that still won't be at full production) meaning you need to push further and futher away from your base. You will need a LOT of rails and you will be thanking your early game self, for automating that production line.
Now its a requirement, which also means Stone now becomes ACTUALLY USEFUL. Before, stone was just something you needed once in a while for walls but mostly just rails. Draining one or two stone locations provided more then enough stone for...well the rest of the game. Now you will need to secure stone spots as well as Iron and Copper spots as the game moves on. Coal usually gets replaced soon and the grenade requirement on military doesn't really take that much coal, so usually its just needed for Plastic. Having a few Coal spots with a few small trains is more then enough to feed that requirement.

Very excited and looking forward to playing .17 once released to beta.

I certainly agree, I always automate rail production. That doesn't mean I'm going to like putting it in a science pack - actually, the opposite as usually I like to have a buffer of them, and science production can very easily drain that buffer, and that means that yeah, I'll be automating the rails going into science (hopefully not because I hope I won't have to), but I'll be automating the rails going into my rail network somewhere else!

[V453000]

...
Are you referring to the catastrophe following FFF#266? I don't like anything in that one, but IMHO, the removal of AM ingredient limits is by far the least offensive of the announced changes. Wube's behaviour following it actually has me warning my friends off this game (that included not updating the roadmap thread for three months and then deleting my post when I did. And that's not the only part of it that is Blizzcon 2018 level bullshit.)

I'm sorry but that's a massive overreactiion.

[Zavian]

- Productivity modules make the factory slower so you either need to make it much bigger or start teching to beacons, as a result require much more power, motivate to use assembling machines 3 because the productivity bonus stacks in the machine and then multiplies between production steps, and makes you able to build much less mining. Also it's the module that snowballs and lets you build the next-tier modules cheaper, that way it's a great idea to start using them from lvl1.

For me, the single biggest reason not to use Productivity modules 1 and 2 is the hassle of upgrading them as the factory advances. Is the integrated version of upgrade planner going to allow me to not just specify "upgrade productivity module 1 to productivity mod 2", but to "set the modules in these machines to 3 productivity module 2s, and one speed module 2." (Or any other combination I might choose). Note that I when I do deploy modules in a non-beaconed factory, I often use 3 prod modules 3 + 1 speed 3, as that gives a decent prod bonus, without needing to massively expand things to compensate for prod modules slowing machines down.

(Another downside to adding prod modules, is that I often use direct insertion, and adding prod modules changes the ratios enough that I would sometimes want to change the design for that section of the factory. It isn't worth doing that more for 3 different module levels).